Standardization of analysis sets for reporting results from ADNI MRI data

Abstract

The ADNI 3D T1-weighted MRI acquisitions provide a rich dataset for developing and testing analysis techniques for extracting structural endpoints. To promote greater rigor in analysis and meaningful comparison of different algorithms, the ADNI MRI Core has created standardized analysis sets of data comprising scans that met minimum quality control requirements. We encourage researchers to test and report their techniques against these data. Standard analysis sets of volumetric scans from ADNI-1 have been created, comprising: screening visits, 1 year completers (subjects who all have screening, 6 and 12 month scans), two year annual completers (screening, 1, and 2 year scans), two year completers (screening, 6 months, 1 year, 18 months (MCI only) and 2 years) and complete visits (screening, 6 months, 1 year, 18 months (MCI only), 2, and 3 year (normal and MCI only) scans). As the ADNI-GO/ADNI-2 data becomes available, updated standard analysis sets will be posted regularly.

A Novel Grading Biomarker for the Prediction of Conversion From Mild Cognitive Impairment to Alzheimer's Disease

“…The data from these studies was released and are also subsets of the standardized set [32]. The same cross validations as in [20] and [22] were used for fair comparisons.…”

“…The standardized list of baseline scans from ADNI-1 [32] was utilized in our work to enable future comparisons with other methods. The standardized list consists of 818 subjects that were classified into five different groups as shown in Table I.…”

“…The main novelties of the proposed MR-based grading biomarker are: (1) a well controlled registration accuracy to align MR images to a template space for comparison; (2) the removal of confounding effects associated with aging; (3) a robust feature selection step which is performed on discriminating NC versus AD, thus eliminating the use of MCI subjects in the feature selection step, and (4) the propagation of disease information from NC and AD subjects to MCI subjects using sparse representation technique. The classification performance of the proposed biomarker was validated on the standardized list [32] of baseline scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and compared with those of other stateof-the-art methods [20], [22], [33] on the same datasets. Furthermore, the proposed grading biomarker is combined with cognitive measures to gain further improvements in the prediction of MCI-to-AD conversion.…”

Intra-Rater and Inter-Rater Reliability of Brain Surface Intensity Model (BSIM)-Based Cortical Thickness Analysis Using 3T MRI

“…The age range of our participants was 55-84, which did not differ from the comparison data. The present study indicated thinner cortices in the entire brain, compared to previously published histological data (8), which may have been induced by use of higher strength MR, a different type of T1 volumetric sequence (SPGR instead of magnetization prepared rapid acquisition gradient echo [MPRAGE]) (29,30), and ethnic or cultural differences in cortical thickness (10)(11)(12). Recruitment of participants from our dementia clinic, which is a university-affiliated hospital, may have resulted in selection bias.…”

“…Although we used an optimized T1-weighted 3D SPGR sequence, this may not be the optimized sequence for cortical thickness analysis by BSIM. Furthermore, the ADNI study recommended a MPRAGE sequence for their multicenter study, due to segmentation power (29,30). Specifically, the ADNI study suggested use of inversion recovery-prepared fast SPGR or MPRAGE sequences for assessing cortical thicknesses (30).…”